ABSTRACT

Introduction:

Mobile health applications (mHealth apps) have the potential to assist patients in adhering to their physician’s advice in chronic disease management through the use of persuasive nudge. However, systematically developing the persuasive features of an mHealth app for the major user demographic of older adults is challenging. The current usability engineering framework could ensure the user-friendliness of the app but not the persuasiveness. It is necessary to extend the current framework with appropriate measures to better understand the effectiveness of persuasive design elements in an iterative design process.

Methods:

A pilot design project was run, a persuasive mHealth app for dietary management was developed using the user-centered design approach (persona, use scenario, task analysis, and cognitive walkthrough), the pilot testing result showed high potential of technology acceptance of older adults. To further evaluate persuasiveness, a food choice experimental protocol and human decision performance metrics based on Signal Detection Theory (SDT) were proposed. A mixed methods, full factorial user testing study was conducted with twenty older adults aged over 60 and twenty students age 18-35. Critical persuasive User Interface (UI) design variables included decision paradigm (digital nudge), nutrition information format (information nudge), and the system default pre-selection (default nudge). The proposed SDT metrics to evaluate persuasiveness were then compared with confusion matrix metrics which are frequently used to validate system decision-making performance. The relationship between the human performance, subjective workload, and perceived usability of the proposed mHealth app was also investigated.

Results: 

The ‘Two Alternative Forced Choice’ layout significantly increased the d-prime and accuracy (persuasiveness), the system default pre-selection decreased persuasiveness. The interpretative FSA Nutri-scores label reduced time of response and workload, and increased perceived ease of use, perceived ease of learning, and satisfaction. Among older adults, results differed by age, computer proficiency, and health literacy. 

Conclusion:

The findings of this study imply the proposed framework is a valid persuasive design research approach. And digital nudge is an effective persuasive design for mHealth app, while default nudge may give rise to negative effects. A generalized human-centered digital nudge design framework along with ageing-centered guidelines were suggested for the similar research and design projects for persuasive technology performed in the future.